Innsworth Capital Limited, R (on the application of) v The Competition Appeal Tribunal & Ors

The claimant, Innsworth Capital Limited, is a litigation funder. It funded collective proceedings brought against Mastercard by Mr Walter Merricks, who acted as the class representative. The proceedings were only the second to be issued after the collective proceedings regime came into effect on 1^st October 2015. After some eight years of litigation, Mr Merricks and Mastercard agreed that Mastercard would pay £200 million to settle the proceedings, a tiny fraction of the sum claimed.

In this claim for judicial review the claimant challenges the order made by the Competition Appeal Tribunal (‘the CAT’) for the distribution of the proceeds of the settlement. The CAT has taken no part in this judicial review, but the challenge to its order is opposed by Mr Merricks. In reality, therefore, the present claim reflects a dispute between Mr Merricks as the class representative and the claimant as the funder as to how much of the settlement proceeds should go to the class and how much to the funder.

The claim against Mastercard was ambitious. It was a follow-on claim for damages allegedly suffered as a result of what the European Commission had found to be unlawful multilateral interchange fees (‘MIFs’) charged to banks on credit and debit card transactions involving banks in different European Economic Area member states between 22^nd May 1992 and 21^st June 2008. The case theory was that everyone in the UK who purchased goods or services during that period from a retail outlet where such cards were accepted had suffered damage in the form of unlawfully inflated charges which the banks had passed on, even where the purchase involved no cross-border element. So the class comprised everyone who was resident in the UK and over the age of 16 at any time between 1992 and 2008 and who purchased goods or services in or from the UK – effectively the entire adult population of the UK during those years – some 44 million people. The total damages were said to amount to about £14 billion including interest up to September 2016 when the proceedings began. As the claim was for compound interest, its value increased as the litigation went on.

The proceedings lasted for over eight years. They raised some tricky points of competition law. There were visits to the Court of Appeal and the Supreme Court. There were many hearings before the CAT. These included a trial of preliminary issues in January 2023 and a three-week trial in July 2023 on issues of (among other things) causation. In January 2024 there was a two-week trial on issues of limitation. The lawyers and the experts did very well out of all this. By the time the case concluded the funder had spent more than £41 million on legal costs and experts’ fees.

The result was that substantial parts of the claim were held to be time-barred and it had become apparent that it would be (at best) very difficult for Mr Merricks to prove that UK consumers had suffered any financial loss as a result of the cross-border MIFs which the Commission had found to be unlawful. So Mr Merricks decided to settle the case and avoid an imminent further trial on an issue on which he was likely to lose. After some negotiations Mastercard agreed to settle the case by paying £200 million. This was only 1.4% of the sum originally claimed and still less if the claim for a further eight years of compound interest is taken into account. Mr Merricks accepted this offer.

Although £200 million is a lot of money, it would not go far if shared equally between the more than 44 million members of the class. In fact, if shared equally between them, with nothing at all for the funder, each class member would receive only about £4.50. Anything paid to the funder would reduce that amount.

The settlement agreement could only take effect if approved by the CAT as ‘just and reasonable’ (Competition Act 1998, section 49A(5)). Approval was resisted by the claimant funder, which wanted to fight on, presumably in the belief that it could get a better deal. But the CAT (Mr Justice Roth as Acting President, Mr Hodge Malek KC and Professor Rachael Mulheron KC (Hon)) approved the settlement. It found that ‘the likelihood of judgment being obtained for an amount significantly in excess of £200 million was low’; that the terms available from Mastercard ‘were the best terms then available’; and that the settlement ‘was in the best interests of the class’. Those conclusions are not now challenged. Nor is the CAT’s description of the outcome as ‘very far from a success for a class of some 44 million claimants’ and as representing ‘an extraordinarily low proportion’ of the claim.

The CAT also had to approve the distribution of the settlement proceeds. That required it to decide how much of the £200 million should go to the class, how much should go to the funder, and what should happen to any money which was not claimed.

The claimant’s primary case before the CAT was that the £200 million should be divided equally between all members of the class and that reimbursement of what it had spent together with its profit should come out of the undistributed damages (i.e. that part of the £200 million which was not taken up by class members). That might seem generous, putting the interests of the class first, but the problem with this approach was that equal division between all class members would result in each member being offered only about £4.50. So very few, if any, would bother to claim. Thus effectively the whole of the £200 million would remain available for distribution to the claimant.

The claimant submitted that, by one route or another, it should be awarded £179 million of the undistributed settlement proceeds. It said that this corresponded to what had been agreed in its Litigation Funding Agreement with Mr Merricks as its agreed minimum return (i.e. a total amount of just under £60 million which the claimant said it had agreed to fund, multiplied by three). Alternatively, it said that this was the market value of the funding commitment which it had provided (i.e. that its terms reflected the litigation funding market). The claimant accepted that, if this approach was followed, any undistributed proceeds out of the residue of £21 million should go to the Access to Justice Foundation, a charity engaged in giving grants to a wide range of organisations providing legal advice or assistance without charge to those in need.

The CAT rejected the claimant’s approach. It decided that the £200 million should be divided into three ‘Pots’. Pot 1, consisting of £100 million, was to go to the class. However, because the equal division of this £100 million between class members would result in minimal take-up, the CAT accepted a proposal by Mr Merricks that each class member taking up the settlement offer should receive a minimum of £45. Statistical advice suggested that this was likely to lead to a take-up by about 5% of the class. The CAT described £45 as ‘an appropriate modest sum given the low aggregate amount obtained for the class, but an amount that is likely to be sufficiently high to attract a likely take-up of at least 5% and thus to be achievable out of the total recovery’. Another way of putting the matter would have been to say that the outcome of the litigation for the class was so poor that 95% of the class members were unlikely to bother to take up the money which was being offered to them.

The CAT recognised that the figure of £45 would have to be reduced if a greater number of class members should submit claims, and that it could be increased if, contrary to expectation, the take-up rate was lower. In the latter event, the CAT imposed a cap of £70 on the amount payable to any individual class member.

The CAT recognised that if 5% of the class did submit claims and were paid £45, that would effectively use up the £100 million in Pot 1. If 10% of the class were to claim, the whole £200 million would be used up, with nothing for the claimant. It regarded that potential outcome as unjust and therefore decided that the amount payable to the claimant (which it then went on to consider) should be ring-fenced so that the claimant’s recovery would not depend on the rate of take-up of the settlement monies by members of the class:

‘130. If, as envisaged, some 5% of the class (i.e. 2.2 million) should submit claims, then at just over £45 each the payments will total £100 million. However, if 10% should claim, then the whole £200 million would be required, with obvious gradation in between. Once this approach is adopted, it would not be appropriate to leave any payment of Innsworth’s return (including reimbursement of its expenditure on legal costs and expert’s fees) to be recovered only out of undistributed damages since the amount of undistributed damages may prove insufficient. Therefore we agree that once this approach is followed, it is necessary and appropriate to reserve a sum out of the £200 million for payment to Innsworth.

131.

That is the basis for the proposal for a pot of £100 million to be reserved for CMs. …’

Pot 2 consisted of the expenditure which the claimant had incurred and had yet to incur in funding the proceedings. The precise figure for Pot 2 could not be determined, but it amounted to at least the £41 million which had already been spent and, taking account of future commitments, might approach £46 million.

That left some £54 million for Pot 3. The CAT decided that the money in this pot should first be used to pay some relatively minor costs and expenses which did not fall within Pot 2. There is no challenge to this aspect of the decision. It should next be used to pay the claimant a just and reasonable profit on its investment. Finally it should be used to supplement Pot 1 in the event that more than 5% of the class submitted claims. If money remained in Pot 3 (i.e. if it was not taken up by class members), it should go to the Access to Justice Foundation.

That left the question, what would be a just and reasonable profit for the claimant? The CAT began its analysis by recognising that a balance had to be struck between the interests of class members and the need to provide fair remuneration for a funder without whose participation the collective proceedings regime could not function effectively:

‘121. We regard it as fundamental that the collective proceedings regime should operate for the benefit of CMs and not primarily for the benefit of lawyers and funders. At the same time, the regime could not function effectively without the CR having good legal representation and commercial litigation funding to pay for it. That presents a particular challenge when, as in the present case, the damages recovered are so very far below the amount envisaged when the proceedings were launched. …’

The CAT noted that it had a broad discretion to determine the funder’s return in the exercise of its wide supervisory jurisdiction over collective proceedings. It identified a number of factors to be taken into account, as follows:

Having identified these factors, the CAT stated its conclusion as follows:

‘188. On that basis, in our judgment a ROI of 1.5 is here appropriate, taking account of all of the above factors, and recognising the significant risk but reflecting also the poor outcome. If the reimbursement of costs and expenses (i.e. Pot 2) should amount to £45.5 million pounds as estimated in the Application, then, applying a ROI of 1.5 will provide Innsworth with a total return of £68 million.’

Thus the profit for the claimant was expected to be about £22.5 million or 11.25% of the settlement proceeds.

It is apparent that this reference to the ROI was a reference to the claimant’s total return (i.e. expenditure plus profit) as a multiple of the expenditure. It was not just referring to the profit. Thus:

Expenditure (£45.5 m) x 1.5 = Total return of £68.2 m

Whether that was a correct understanding of the Australian case law to which the CAT referred is one of the issues arising on this judicial review.

It is clear that the CAT regarded the outcome of the litigation as ‘poor’ or ‘very poor’ for the class. This is a view which has not been challenged in these proceedings. It is a view which the CAT was entitled to reach. It was clearly the principal driver of the CAT’s conclusion that the claimant’s profit should not exceed 50% of the expenditure which it had incurred. The CAT made this point repeatedly, not only in the passages to which I have already drawn attention, but elsewhere also (my emphasis):

‘184. Mr Béar relied on the passages in the Street judgment on the danger of engaging in hindsight bias when assessing a funder’s return. The need to avoid such bias was indeed expressed by the full Federal Court of Australia in Money Max: para 170 above. But at the same time the return cannot disregard the degree of success or failure of the proceedings. We see nothing in Murphy J’s judgment in Street to weaken the observations made by the same judge in Petersen, when scrutinising the proposed settlement of a class action arising out of a fraudulent Ponzi scheme. He there said, at [5]-[6]:

“The Part IVA regime is intended to provide access to justice to the applicant and class members and it is not intended solely for the benefit of service providers such as lawyers and funders. The legitimate use of the Court’s processes should not be undermined by proceedings that disproportionately benefit the funder and/or solicitor rather than the litigants.

For the reasons I explain, in the particular circumstances of the case, it is appropriate to approve the settlement but to disallow a substantial amount of the legal costs and funding charges sought”.’

The CAT returned to this point at the end of its judgment. It explained that the distribution which it had ordered would not set a precedent in all cases. Rather, the poor outcome in the present case did not reflect the public policy purposes of collective proceedings and the return for the funder had to recognise this:

‘208. We emphasise that the approach to settlement set out in this judgment is determined by the exceptional circumstances of this case, where the settlement is at an extraordinarily low proportion (under 1.5%) of the claim as originally advanced in 2016, and under 1.2% of the claim as revised in late 2022. This approach should not be regarded as a guide for more positive settlements of cases that reflect better the public policy behind the introduction of collective proceedings.’

Having reached its conclusion that the appropriate level of profit for the claimant was 50% of the expenditure which it had incurred, the CAT applied what it called ‘a useful cross-check’:

‘188. … Although as the law stands funders cannot conclude a LFA [Litigation Funding Agreement] providing for a percentage-based return, and we therefore have not determined the return on that basis, it nonetheless serves as a useful cross-check. A payment of £68 million constitutes 34% of the Settlement Sum. That contrasts with the much lower percentage of 16% determined by Murphy J in Street (where the circumstances were exceptional), and is within the “presumptive range of validity” established by the Canadian jurisprudence: para 175 above.’

In saying that this was the position ‘as the law stands’, the CAT had in mind the decision of the Supreme Court in R (PACCAR Inc) v Competition Appeal Tribunal [2023] UKSC 28, [2023] 1 WLR 2594.

It might be thought that a guaranteed total return of about £68 million, representing a profit of 50% on the claimant’s investment, was not a bad result for the claimant. If the case had not settled, it would not only have borne the cost of another expensive trial, but would probably have lost its entire investment and made no profit at all. As the CAT observed, Mr Merricks’ remaining argument in the claim against Mastercard ‘faced very substantial obstacles’. But the claimant is dissatisfied with its allocation of the settlement proceeds. It challenges this allocation, contending that it involved several errors of law by the CAT. It has to do so by way of judicial review because, although the usual route for challenging a decision of the CAT is by an appeal to the Court of Appeal, that route is not available to a funder which has participated only as an intervenor in the proceedings before the CAT.

The decisions which the claimant seeks to challenge are pleaded as ‘the decisions of the CAT in relation to Pot 1 and Pot 3’. In other words, it does not challenge the CAT’s decision on Pot 2 (i.e. that it should be reimbursed for its expenditure and that such reimbursement should be ring-fenced). It confines its challenge to what the CAT decided about Pot 1 and Pot 3. However, the CAT’s decision in relation to all three Pots was a single package. Its rationale for ring-fencing reimbursement of the claimant’s expenditure (Pot 2) was inextricably linked to the way in which it dealt with Pots 1 and 3. That is clear, for example, from what the CAT said at paras 130 and 131 of its judgment which I have quoted above.

This led Mr Mark Brealey KC for Mr Merricks to submit that the claimant is challenging the wrong decision; that there has been no challenge to the CAT’s actual decision, which was the package as a whole; and that the claim for judicial review must therefore fail on that ground alone. I would be reluctant to dismiss the claim on that pleading ground if the claimant’s challenge to the CAT’s decision is otherwise well-founded. Rather, if the claim were to succeed, the CAT’s decision as a whole would have to be quashed.

The application for permission to bring a claim for judicial review came before Mr Justice Sheldon on paper. He decided to adjourn the application to an oral hearing, which came before Mr Justice Linden.

After an oral hearing Mr Justice Linden gave permission to bring a claim for judicial review on three grounds, as follows:

Mr Justice Linden refused permission on other grounds which the claimant had sought to argue. It now seeks to renew its application for permission in relation to two of those grounds, namely:

Time did not allow proper consideration of these two grounds before the hearing. We therefore directed that the renewed application for permission should be dealt with by way of a rolled-up hearing together with the grounds for which permission had been given.

It is worth saying that although the hearing before the CAT lasted three days, the first two of those days were concerned with the issue whether settlement of the claim against Mastercard for £200 million was just and reasonable, an issue which no longer arises. The third day was concerned with the distribution of the proceeds of the settlement, but the principal issue which the CAT had to decide was whether (as the claimant proposed) the £200 million should be divided equally between all members of the class with the claimant’s return coming out of the undistributed damages or (as Mr Merricks proposed) the approach of separate pots of money should be adopted. Although the issues currently raised about what should go into the various pots were argued before the CAT, it is clear that the arguments about this in these judicial review proceedings have been much fuller than the arguments before the CAT. In those circumstances I would pay tribute to the clear and full consideration of those arguments in the CAT’s judgment ([2025] CAT 28).

Section 49A(1) and (2) of the Competition Act 1998 empower the CAT to approve the settlement of claims in collective proceedings, and provide that an application for approval of such a settlement must be made by the class representative and the defendant. Section 49A(5) then provides that:

‘The Tribunal may make an order approving a proposed collective settlement only if satisfied that its terms are just and reasonable.’

Rule 94 of the Tribunal Rules provides that ‘opt-out’ collective proceedings (meaning collective proceedings brought on behalf of class members except those who opt-out within the meaning of section 47B(11) Competition Act 1998) may not be settled other than by a collective settlement approval order issued by the CAT in accordance with this rule. An application for such an order must, among other things, ‘specify how any sums received under the collective settlement are to be paid and distributed’ (para (4)(d)).

The CAT held that it is implicit in this Rule 94(4)(d) that not only the terms of the settlement, but also the arrangements for distributing the settlement proceeds, require the approval of the CAT, and that the question for the CAT is whether those arrangements are just and reasonable. That conclusion has not been challenged before us and is clearly correct. As Lord Briggs put it when this case went to the Supreme Court at the certification stage ([2020] UKSC 51, [2021] Bus LR 25, para 58):

‘The only requirement, implied because distribution is judicially supervised, is that it should be just, in the sense of being fair and reasonable.’

This comment refers to distribution when there has been an award of damages, but applies equally when the claim has been concluded by a settlement. In deciding how the proceeds should be distributed, the CAT is exercising judgment as an expert specialist body, taking account of a range of factors. It does so pursuant to a regime which ‘is in many respects a departure from established tenets of civil liability’(Le Patourel v BT Group Plc [2022] EWCA Civ 593, [2022] Bus LR 660, para 89), and in the exercise of ‘wide unrestricted powers’ (Gutmann v Apple Inc [2025] EWCA Civ 459, [2025] Bus LR 1292, para 78).

Accordingly the CAT has a very wide scope to decide how the proceeds of settlement should be distributed as between the funder and the members of the class. This has sometimes been described as an exercise of discretion. It may be better viewed as an evaluative assessment, although for practical purposes this will make little if any difference.

The important point for present purposes is that the CAT’s decision will only be open to challenge on appeal if there is some identifiable flaw in its treatment of the question to be decided, such as a gap in logic, a lack of consistency, or a failure to take account of some material factor, which undermines the cogency of its conclusion (In re Sprintroom Ltd [2019] EWCA Civ 932, [2019] BCC 1031, paras 76 to 78). Thus it is not any error which requires a decision of the CAT to be set aside on appeal. The error must be material, in the sense that it is sufficiently important to undermine the cogency of the CAT’s conclusion.

Accordingly an appellate court would need to be cautious before interfering with the decision of a specialist tribunal such as the CAT on a question whether the arrangements for distributing the proceeds of a settlement are just and reasonable. In the case of an appeal from the CAT, a further limiting factor is that an appeal will only lie on a point of law, although it has been said that this requirement should be construed broadly (Competition Act 1998, section 49(1A)(a) and O’Higgins v Barclays Bank Plc [2023] EWCA Civ 876, [2024] Bus LR 366, paras 44 to 47 and 57).

The limited scope of an appeal from an evaluative judgment of the CAT was emphasised in Evans v Barclays Bank Plc [2025] UKSC 48, [2026] Bus LR 328 (on appeal from O’Higgins):

‘128. In our view, the essential error made by the Court of Appeal in its analysis of practicability was that it failed to follow the sound guidance given in its earlier judgment in Le Patourel about the approach the Court of Appeal should adopt when considering evaluative judgments made by the Tribunal and instead made its own assessment of the evidence, substituting its own evaluation for that of the Tribunal.

129.

In Le Patourel, at para 57, the Court of Appeal considered the circumstances in which it is legitimate for the Court of Appeal to interfere with the evaluation of facts made by the Tribunal. Green LJ said that:

“when it comes to the weighing up of the various factors relevant to the choice of opt-out or opt-in this is essentially an exercise of judgment over facts and evidence by an expert, specialist, body, that will over time accrue an increasing well of experience in how to handle these complex cases. The appellate courts recognise that the case management decisions of the [Tribunal] are exercises in pragmatism and that undue formalism and precision are not required … These considerations broaden the Tribunal’s margin of discretion or judgment. This court should not interfere simply because it might, for the sake of argument, have drawn a different conclusion from the weighing exercise. We would expect that most opt-out/opt-in decisions will involve a weighing exercise of this nature.”

130.

In Le Patourel the Court of Appeal also pointed out, at para 83, that the concept of practicability is “highly fact and context sensitive and will include, as one facet of this broad analysis, the Tribunal using its expert judgement to envisage how the costs and benefits of litigation will play out upon an opt-out or opt-in basis. The view the Tribunal takes will be a conclusion reached by an expert body with a growing depth of experience in the conduct of collective proceedings”. In that case the Tribunal had examined relevant factors such as size of class, the scale of a possible award and the impact of these on funding as important considerations. The Court of Appeal concluded, at para 74:

“The [Tribunal] came to specialist conclusions which lay squarely within its broad margin of judgement. There is in our judgment no basis in law upon which this court can properly interfere”.’

Although the context for these remarks is whether to certify collective proceedings as opt-in or opt-out, they are equally applicable to the question whether a decision as to the distribution of settlement proceeds is just and reasonable.

Of course, this is not an appeal, but a claim for judicial review. Accordingly Mr Charles Béar KC for the claimant invoked principles of public law, including the twin concepts of process rationality and outcome rationality explained by Mr Justice Chamberlain in R (KP)v Secretary of State for Foreign, Commonwealth and Development Affairs [2025] EWHC 370 (Admin), paras 56 and 57, and reiterated in R (BEL) v Secretary of State for Foreign, Commonwealth and Development Affairs [2025] EWHC 1970 (Admin) para 108.

It may be that it does not matter much, if at all, which of these approaches is followed. As Lord Justice Green observed in O’Higgins, para 58, in practical terms there is not a great deal of difference between an appeal on a point of law and judicial review. However, if and to the extent that it does matter, it would seem to me to be perverse to allow a non-party to the collective proceedings a greater scope for challenging the decision of the CAT by way of judicial review than would be possible on an appeal on a question of law by a party to the proceedings. Accordingly, in considering the lawfulness, including the rationality, of the CAT’s decision for public law purposes, we should adopt the same approach as would apply on an appeal.

As I have explained, one of the factors which the CAT took into account in arriving at the appropriate profit for the claimant was evidence from Australian case law which suggested that an ROI of between 1.2 and 1.9 was typical for one substantial litigation funder. Mr Béar submitted that the CAT misunderstood this case law, because the ROI referred to in the Australian cases was not what I shall call the total return (i.e. expenditure plus profit) expressed as a multiple of the expenditure, but the profit element only.

The case to which the CAT referred was Street v State of Western Australia [2024] FCA 1368, a decision of Justice Murphy in the Federal Court of Australia. It was an application for court approval of a settlement of a claim made on behalf of Australian First Nations peoples for widespread non-payment and underpayment of wages between 1936 and 1972. Justice Murphy regarded the claim as one which faced substantial legal hurdles and substantial risks. On that basis, he approved a settlement of AU$159 million as fair and reasonable, even though this represented a very substantial discount on full compensation.

Dealing with the profit (referred to as the funding commission) for the funder to be paid out of the settlement proceeds, Justice Murphy said this:

‘358. Assuming there are 8,750 OECs [Original Eligible Claimants] the gross settlement amount is $159,775,000 million … and the Funder seeks a 20% funding commission being $31,955,000.

359.

That funding commission, combined with reimbursement of the legal costs and ATE Costs the Funder paid ($14,403,868), would mean that the Funder would receive $46,358,868 million from the proposed settlement, based on an investment of $14.403 million … That would give the Funder an ROI of approximately 3.22 times …’

Justice Murphy determined that an ROI of 3.22 was too high:

‘361. In the circumstances of the case, particularly the quantum of the settlement and the way the funding terms operated in practice, I do not consider a funding rate of 20% representing a commission of almost $32 million and at a 3.22 times ROI to be fair and reasonable. I consider a funding rate of 16% … to be commercially realistic and to properly reflect the costs and risks taken on by the Funder. Money Max at [82].

362.

I accept that a 20% headline funding rate is toward the bottom of the range of the rates available on the market and, intuitively, a funding rate of 16% seems too low. However, having regard to the quantum of the settlement, the costs and risks the Funder took on, and the operation of the funding terms in the circumstances of the case I consider that funding rate of 16% to be “just”. In particular:

…

(g)

that funding rate provides the Funder an ROI of 2.77. That is not a niggardly return. Recently, in Allen & Anor v GB Education Ltd (No 4) [2024] VSC 487 at [110] Watson J noted the following:

“Omni Bridgeway, an ASX listed litigation funder, publishes data regarding its MOIC which is the total amount it receives (including an return of its investment amount) divided by the amount invested (but not including finance costs). In other words, the MOIC is the ROI plus one if finance costs are excluded. The evidence shows that Omni Bridgeway’s ROI on all completed cases (including those on which it loses some or all of its capital) is 1.2 and approximately 1.9 on those cases which did not produce a negative return. Approximately 15% of its cases have an ROI exceeding 4.0 with some cases having an ROI exceeding 9.0.”

363.

A 16% funding rate means (on the assumption of 8,750 OECs) that the Funder will receive a total of $39,967,868 (representing a funding commission of $25,564,000 and the reimbursement of its investment of $14,403,868).’

It is crystal clear from the arithmetic in these passages, that the ROI of 2.77 (‘not a niggardly return’) represented the total return (expenditure plus profit) expressed as a multiple of the funder’s expenditure. In round figures, expenditure of $14 million plus profit of $25.5 million amounted to a total return of $39.5 million, which was 2.77 times the funder’s expenditure of $14 million. Expressing the profit alone as a multiple of expenditure would have resulted in a figure of 1.77.

In contrast, in the earlier Australian case of Allen v G8 Education Ltd (No. 4) [2024] VSC 487 cited by Justice Murphy, it appears that ROI was used in a different sense, and did express the profit alone as a multiple of the expenditure: hence the comment in Allen that ‘the MOIC [multiple on invested capital] is the ROI plus one’. This was made explicit at para 78, which explained that:

‘The ROI (return on investment) is simply the profit or return from the investment divided by the cost of the investment.’

Although this definition of ROI was repeated in Street at para 269, and although I hesitate to say so in view of the fact that Justice Murphy is evidently very experienced in dealing with this kind of case, it appears from the figures set out above that he did not use the term consistently. He said that:

‘269. A payment of $46,358,868 for an investment of $14,403,868 would provide an ROI of 3.22 times (although that does not include adverse costs risk). By ROI (return on investment) I mean the profit or return for the Funder from the investment divided by the cost of the investment.’

Although the explanation of what is meant by ROI appears to equate return with profit, it is apparent from the figures that the ROI of 3.22 to which Justice Murphy referred was the total return as a multiple of the investment (Investment of c.$14 million x 3.22 = Return of c.$46 million). Similarly, when he came to the critical paragraphs which I have set out above, the figures demonstrate that he was using ROI to refer to total return expressed as a multiple of expenditure.

It was Allen which set out the evidence showing an ROI of between 1.2 and 1.9 for what the CAT described as ‘a substantial litigation funder’ (see para 17(5) above). The full passage is as follows:

‘108. In its materials Slater and Gordon compares its ROI to the multiples which third party litigation funders seek and to Omni Bridgeway’s published multiple on invested capital (‘MOIC’).

109.

Litigation funders generally seek funding opportunities at a multiple of at least two or three times, equating to an ROI of 2.0 or 3.0. Actual ROI’s for litigation funders are presumably less, because some cases will not produce those returns but nonetheless it provides a point of comparison for Slater and Gordon’s ROI in this proceeding. Slater and Gordon’s calculated range for its ROI for this is reasonable and appropriate having regard to that evidence.

110.

Omni Bridgeway, an ASX listed litigation funder, publishes data regarding its MOIC which is the total amount it receives (including any return of its investment amount) divided by the amount invested (but not including finance costs). In other words, the MOIC is the ROI plus one if finance costs are excluded. The evidence shows that Omni Bridgeway’s ROI on all completed cases (including those on which it loses some or all of its capital) is 1.2 and approximately 1.9 on those cases which did not produce a negative return. Approximately 15% of its cases have an ROI exceeding 4.0, with some cases having an ROI exceeding 9.0. Slater and Gordon’s calculated range for its ROI excluding finance costs is reasonable and appropriate having regard to that evidence.’

It appears, therefore, that the term ROI is not used consistently in the Australian cases. In Allen it always refers to the funder’s profit expressed as a multiple of expenditure. In Street it is sometimes described in this way, but in fact the sum awarded, also described as an ROI, is the total return (expenditure plus profit) expressed as a multiple of expenditure – what in Allen was referred to as the MOIC (or ROI +1).

What the CAT drew from these cases was that an ROI of between 1.2 and 1.9 on all completed cases experienced by a substantial litigation funder was a relevant metric in deciding what would be an appropriate profit for the claimant in the present case. It is clear from Allen, in which these figures are referred to, that they represent profit (and not expenditure plus profit) expressed as a multiple of expenditure. If the CAT understood these figures as referring to total return (expenditure plus profit) expressed as a multiple of expenditure, as seems likely, it was mistaken.

In his reply submissions Mr Béar sought to put before us the published data from Omni Bridgeway referred to in Allen, submitting that this confirmed that the ROI referred to was indeed the profit expressed as a multiple of expenditure. This provoked a vigorous debate in post hearing submissions as to what the data actually shows. In the event it is unnecessary to consider this material, because the position is clear from the judgment itself. There is no getting away from the formula that ‘MOIC is the ROI plus one’. Had it been necessary, however, I would not have been prepared to admit this evidence, which came far too late and was not before the CAT.

In my view, however, the CAT’s misunderstanding does not come close to undermining the cogency of its conclusion that a profit of 50% of the claimant’s expenditure would be just and reasonable. As the CAT made clear, its conclusion was based on all of the factors which it had identified, and not on the Australian evidence alone. I set out again the relevant paragraph of the judgment, with added emphasis:

‘188. On that basis, in our judgment a ROI of 1.5 is here appropriate, taking account of all the above factors and recognising the significant risk but reflecting also the poor outcome. If the reimbursement of costs and expenses (i.e. Pot 2) should amounted to £45.5 million as estimated in the Application, then, applying a ROI of 1.5 will provide Innsworth with a total return of £68 million. …’

It is clear, therefore, that the Australian evidence was only one factor among several to which the CAT had regard. Far more important than the Australian evidence were the poor outcome of the proceedings for the class, the fundamental principle that ‘the collective proceedings regime should operate for the benefit of [class members] and not primarily for the benefit of lawyers and funders’ (para 121), and the need to ensure that what is recovered by a funder is not excessive (Gutmann v Apple Inc, para 81). All of these were highly relevant considerations which pointed strongly to the overall justice and reasonableness of the order made by the CAT, including as one component of its order that the claimant would earn a profit of 50% of its expenditure. The CAT made clear that it fully recognised the importance of litigation funding to the claim (para 176), but to have awarded the claimant a profit materially in excess of 50% of its expenditure would have subverted these important principles.

Moreover, the ring-fencing of the claimant’s return, including both its reimbursement of expenditure and its profit, was a critical and valuable aspect of the package which the CAT awarded. Typically, a funder’s return will be paid out of the damages or settlement proceeds which are not distributed to class members, leaving the funder to bear the risk that the level of take-up by class members will leave it with insufficient funds remaining. Indeed, that was the regime for which the claimant contended. But in the present case, that risk was removed. Accordingly the package awarded by the CAT was not directly comparable with the typical case.

For these reasons I would reject this ground of challenge.

Alternatively, and if necessary, I would refuse to grant relief on this application pursuant to section 31(2A) of the Senior Courts Act 1981. This provides:

‘The High Court—

(a)

must refuse to grant relief on an application for judicial review, and

(b)

may not make an award under subsection (4) on such an application,

if it appears to the court to be highly likely that the outcome for the applicant would not have been substantially different if the conduct complained of had not occurred.’

It appears to me to be highly likely, indeed almost inevitable, that the outcome in this case would not have been substantially different if the CAT had correctly understood the effect of the evidence in the Australian cases or if it had not referred to that material at all. The factors to which I have already referred point clearly to that conclusion. Indeed, as I understood his submissions, Mr Béar was inclined to accept, subject to his other grounds to which I shall come, that a process of reasoning as straightforward as saying that the case had turned out badly and that the funder should recover what it had paid together with a profit of 50% could not have been challenged as being unfair or unreasonable.

The case advanced (or which appeared to be advanced) in the claimant’s Statement of Facts and Grounds was that the CAT’s starting point should have been to identify the net proceeds (i.e. £154 million, being the gross settlement proceeds of £200 million less the claimant’s expenditure); it should have awarded reimbursement of expenditure as its first step; and only then should it have considered how to divide the net proceeds between the class and the funder. I would have rejected this submission because it imposes an inappropriate straitjacket on the CAT’s determination of what is just and reasonable, contrary to the ‘wide unrestricted powers’ (see above) conferred on it.

In the event, however, Mr Béar made clear that this was not the claimant’s submission. Rather, it was that the CAT was required to have regard to the net proceeds of the settlement as a factor to be taken into account in determining the allocation of the settlement proceeds, and that it had failed to do so.

I would reject this submission. It is obvious that the CAT was well aware of the expenditure incurred by the claimant. It was well aware also that the net proceeds for division between the class members and the claimant after reimbursement of the claimant’s expenditure were of the order of £154 million. The fact that the CAT began by awarding £100 million to the class and referred to this as Pot 1 (rather than Pot 2) does not begin to suggest that it failed to have regard to the net proceeds of the settlement.

Rather, the CAT took the view that reimbursement of the claimant’s expenditure (Pot 2) plus a profit of 50% of that expenditure (to be drawn from what it called Pot 3) represented a fair and reasonable return for the claimant in view of the ‘very poor result’ of the case; that anything more would not have been appropriate; and that the balance of Pot 3 should be available to class members. That was a conclusion which it was entitled to reach. It was consistent with the objective of the collective proceedings regime, which is to facilitate access to justice for those (in particular consumers) who would otherwise not be able to access legal redress (Le Patourel v BT Group Inc, para 29); with its view that ‘the return cannot disregard the degree of success or failure of the proceedings’ (para 184); and with the need to ensure that what is recovered by a funder is not excessive (Gutmann v Apple Inc, para 81).

When the claim against Mastercard began in 2016, the proceedings were not funded by the claimant but by another litigation funder called Colfax. Colfax terminated its funding arrangements with Mr Merricks in July 2017 after the CAT had refused to certify the proceedings as collective proceedings, a decision later reversed on appeal. The claimant then stepped in to replace Colfax. Despite the termination of its funding agreement, Colfax maintained that it would be entitled to share in any proceeds of the litigation. Mr Merricks denied that Colfax retained any such entitlement.

Without asking or informing Mr Merricks, the claimant entered into an agreement with Colfax dated 4^th February 2021, by which Colfax released Mr Merricks from any rights it had against him under its original funding agreement and assigned any such rights to the claimant. Under this agreement the claimant has an obligation to pay a specified sum of money to Colfax out of any profit it receives from the litigation after reimbursement of its expenditure.

The claimant’s principal submission to the CAT, as already explained, was that there should be an equal division of the settlement proceeds between all the class members and that its return (reimbursement of expenses plus profit) should be paid out of the undistributed proceeds of the settlement. If that approach was adopted, it was content to pay whatever it was obliged to pay to Colfax out of its profit. However, if the CAT adopted the approach for which Mr Merricks contended, as it did, the claimant suggested that it should be paid its liability to Colfax out of Pot 3, in addition to its profit. The CAT rejected this suggestion (the CAT’s emphasis):

‘192. … On [the claimant’s] evidence quoted above, the liability Innsworth assumed to Colfax is to be discharged out of the profit return Innsworth receives. In effect, this amounts to an agreed but limited sharing of that profit return. …’

Mr Béar submitted that the CAT was wrong to characterise the claimant’s liability to Colfax as a profit share. Rather, it was an expense incurred for the benefit of the class, which released the class from what was at least a potential liability to Colfax. The claimant should therefore have been entitled to recover what it was obliged to pay Colfax out of Pot 3 before any of the funds in Pot 3 were made available to the class.

I would reject that submission. The claimant chose to enter into its agreement with Colfax without reference to Mr Merricks. The CAT was right, or at any rate was entitled, to characterise this agreement as an agreed sharing of any profit which the claimant would obtain. There was no error of law in holding the claimant to that agreement and it was not irrational to do so.

As I have explained, the claimant seeks to renew its application for permission to bring its claim for judicial review on two of the grounds for which Mr Justice Linden refused permission at an oral hearing. Mr Brealey, supported by Mr Gerard Rothschild for the Access to Justice Foundation, opposed this application on both procedural and substantive grounds. As to the procedure, they submitted that any application should have been made to the Court of Appeal, referring to CPR 52.8:

‘(1) Where permission to apply for judicial review has been refused at a hearing in the High Court, an application for permission to appeal may be made to the Court of Appeal except where precluded by section 18(1) of the Senior Courts Act 1981.

…

(3)

An application under paragraph (1) must be made within 7 days of the decision of the High Court to refuse to give permission to apply for judicial review.

…

(5)

On an application under paragraph (1) … the Court of Appeal may, instead of giving permission to appeal, give permission to apply for judicial review.’

The claimant sought (and was refused) an extension of the seven day deadline for making an application to the Court of Appeal from Mr Justice Linden, but did not then make any application to the Court of Appeal. Instead, on 12^th March 2026, 31 days after the permission hearing, it made a renewed application for permission to this court on two grounds, relying on CPR 54.15:

‘The court’s permission is required if a claimant seeks to rely on grounds other than those for which he has been given permission to proceed.’

A note in the White Book (Civil Procedure (2026), para 54.15.1) suggests that this is possible in what are described as exceptional circumstances:

‘Where permission to argue a particular ground has been expressly refused at an oral permission hearing, a claimant wishing to challenge that refusal should normally do so by appealing against the refusal to the Court of Appeal. Exceptionally, however, the court hearing the substantive application of the grounds where permission was given, may exercise its discretion under CPR r.54.15 to permit the claimant to argue the specific ground in relation to which permission was refused if there has been a significant change of circumstances, the claimant has become aware of significant new facts which they could not reasonably have known or found out about at the time of the permission hearing, if a proposition of law is now maintainable which was not previously open to the claimant (for example, where the Court of Appeal has overturned a decision of the High Court), or where there is some other substantial justification for allowing the point to be argued.’

The principal case cited in support is R (Smith) v Parole Board [2023] EWCA Civ 1014, [2003] 1 WLR 2548, where Lord Woolf CJ referred to the discretion in somewhat wider terms:

‘16. … Of course, where, as here, a judge has heard detailed argument, any judge who is conducting the hearing of the main application is going to require significant justification before taking a different view from the judge who granted permission. However, if he comes to the conclusion that there is good reason to allow argument on an additional ground, bearing in mind the interests of the defendant, the judge can give permission for that to happen. It is not unusual for a situation to arise, even in the course of the hearing, where it becomes apparent to the judge conducting that hearing that the interests of justice would be best served by the hearing taking into account arguments on matters which relate to a ground in respect of which permission has been refused. There obviously has to be real justification for permitting that to happen; but judges can be relied upon to ensure that the discretion is not misused. … I would not seek to anticipate all situations where that could happen. As long as a judge recognises the need for there to be good reason for altering the view of the single judge taken at the permission stage, no further sensible guidance can be provided. The circumstances which can occur are capable of varying almost without limit, and so each case must be considered having regard to its circumstances. The idea that there has to be a new situation for the permission to be extended as one which I would regard as wrong.’

In the present case permission to bring a claim for judicial review on grounds 1, 2(a) and 4 was granted; the substantive claim was expedited; and the case was directed to be heard by a Divisional Court. It would have been possible for the claimant to request that a single judge of the Court of Appeal should consider an application for permission to appeal urgently, with a view to giving a direction under CPR 52.8(5) that the grounds on which permission had been refused should be argued before the Divisional Court. But that does not strike me as an efficient use of judicial resources. In the unusual circumstances of this case, therefore, I would not have refused the application for permission on the basis that the application should have been made to the Court of Appeal.

Mr Béar submitted that the CAT failed to have regard to the market value of the services provided by the claimant under its Litigation Funding Agreement, referring to principles of unjust enrichment. He submitted that the market standard for litigation funding profit in opt-out collective proceedings at the time of the Agreement was the higher of 30% of total damages or three times the capital committed (i.e. £179 million).

Mr Justice Linden held that this ground was not arguable. In reality it was a repetition of ground 2(b), for which permission was refused, that the claimant was entitled to a minimum floor return provided for in its Litigation Funding Agreement. I agree with that view.

In any event the CAT did consider this submission. It said in terms that it did not regard the principles of unjust enrichment and quantum merit as relevant to its decision (para 186) and, as already explained, it emphasised the importance of having regard to the degree of success or failure of the proceedings. That was a legitimate approach for all the reasons already discussed.

The final ground which the claimant seeks to argue is that the CAT failed to appreciate the basic distinction between (i) allocation as between class and funder, and (ii) allocation as between funder and a windfall recipient such as a charity. Mr Béar submitted that there should be no provision for payment to a charity which had made no contribution to the claim and that any residue should go to the funder, on the ground that it was only because of the funder’s expenditure that there were any settlement proceeds to distribute.

Mr Justice Linden considered that this ground added nothing to the claimant’s other grounds. As he put it:

‘47. I do not see how this Ground adds to the grounds which challenge the profit return which the CAT awarded to Innsworth. If those grounds are sound, the claim may succeed, but not otherwise. The approach of the CAT was separately to determine the just and reasonable figure which should be ringfenced to Innsworth, applying the correct principles, including that what is recovered by the funder should not be excessive. It then awarded the sum which it arrived at, in priority not only over any further payment which may need to be made to the Class in the event of a higher than expected take-up, but also to any payments which might go to the Foundation.

48.

There is no sign in the CAT’s reasoning that its view that any residual sums should go to charity influenced its decision as to what Innsworth should be awarded. It did not, for example, adopt a conservative approach to Innsworth’s profit return so that there would be funds left over for charity. The CAT did not award £30 million to charity: it awarded whatever was left over from pots 1 and 3. Nor was its award at the expense of Innsworth.’

This is in my view a complete answer to the point.

In summary, the CAT determined that despite the very poor outcome of the claim against Mastercard, a just and reasonable return for the claimant funder would be reimbursement of its expenditure of between £41 million and £46 million, together with a profit of 50% of that expenditure. This return was guaranteed, in that it will not depend on the level of take-up of the settlement proceeds by members of the class. The CAT determined also, at least by necessary implication, that any greater return would have been excessive. These were conclusions which it was entitled to reach and which were well within the wide powers conferred upon it as an expert and specialist tribunal.

I would refuse the claimant permission to argue grounds 2(c) and 2(e). I would dismiss the claim for judicial review on grounds 1, 2(a) and 4 for which permission was granted.

I agree.